Objective: To identify risk factors for preterm birth and determine if these vary by degree of prematurity.
Methods: We used data from the state-wide mandatory surveillance system for all births in New South Wales, limiting analysis to the 836,292 live born, singleton infants without known birth anomaly born from 1994 to 2004 inclusive. Our main outcome measure was gestational age stratified into the clinically relevant groups of: ‘term’ (37–42 completed weeks gestation); ‘mildly preterm’ (33–36 weeks); ‘very preterm’ (29–32 weeks); and ‘extremely preterm’ (23–28 weeks). Analysis was by multivariate modelling using a generalised estimating equations model and confidence intervals adjusted to account for the multiple comparisons.
Results: Increasing socioeconomic disadvantage was associated with increasing risk of having a preterm baby. This association strengthened with increasing degree of preterm birth, (adjusted Odds Ratio for mothers from the most disadvantaged areas having an ‘extremely preterm’ baby = 1.45 [99.67% CI 1.21–1.75] compared to least disadvantaged areas). Mothers who were older, who smoked, were Aboriginal, or had pre-existing diabetes, hypertension, or pre-eclampsia were independently more likely to have a preterm baby. First-time mothers were more likely to have their baby at term.
Conclusions and implications: While risk factors for preterm birth such as pre-existing medical conditions are treatable, reducing the substantial effects of socioeconomic factors on preterm birth presents the greatest potential for change. Our data shows that tackling wider social issues will be necessary to assist in reducing the rising preterm birth rate.